Zipher has saved us money. It's mostly autonomous, and can, if enabled across the board, address cost issues almost as quickly as they arise. We were somewhat reluctant and didn't enable it as quickly as we planned (according to Zipher's recommendations), so we left money on the table. Still, we've been implementing Zipher (too slowly) for about 6 months, and we've saved over $1M so far with an estimated $2M annually as we continue to roll it out. Review collected by and hosted on G2.com.

Rarely, but occasionally, we have had some SLA impacts, but Zipher adapted to that and provided a way to put some guardrails in place for that. Also, occasionally, we have forgotten to restart learning after doing manual performance changes, and that also created some minor confusion. Review collected by and hosted on G2.com.

Zipher seamlessly cut our Yotpo data load costs by ~30% through intelligent cluster sizing and parameter optimization—all without sacrificing control. We retain full flexibility to balance cost versus performance and define strict job completion time targets. Additionally, its automated fallback mechanism gracefully deactivates Zipher after a few failed attempts to protect our infrastructure, and full UI and Terraform integration makes management seamless. Review collected by and hosted on G2.com.

Zipher performs exceptionally well for Databricks, so naturally, our main feedback is that we simply want to use it in more places. It would be extremely valuable to see Zipher expand its intelligent cost-saving capabilities to cover general-purpose compute and additional data platform providers beyond Databricks. Review collected by and hosted on G2.com.

I appreciate that Zipher minimizes the operational load on teams to optimize their jobs. Manual tuning can be so time consuming, and is not always practical with a large number of data pipelines. Additionally it adjusts cluster configurations based on time of day which can capture savings you couldn't yourself with a static job configuration. Review collected by and hosted on G2.com.

There have been some early limitations with configuration options and external integration, but those are rapidly improving. Review collected by and hosted on G2.com.

What I appreciate most about Zipher is how easy it is to see tangible results. In my role, I was deeply involved in optimizing cloud costs and searching for tools that could help lower our expenses with both cloud providers and vendors. Zipher’s solution was not only straightforward to integrate, requiring minimal adjustments on our end, but it also delivered a significant impact on our costs. Review collected by and hosted on G2.com.

The main drawback of Zipher is that its cost reduction features are limited to just one area of Databricks—specifically, jobs—while the other aspects are not addressed and require other tools. Review collected by and hosted on G2.com.

Truly hands-off. We cut our Databricks bill without having to touch any code. Review collected by and hosted on G2.com.

Honestly not much. UI could look a bit cleaner, but that's it. Review collected by and hosted on G2.com.

Set it up once and forgot about it - costs dropped and nothing broke. Review collected by and hosted on G2.com.

Honestly not much. UI could use some refining, but that's it. Review collected by and hosted on G2.com.

The spark-aware autoscaler is significantly less wasteful than the native autoscaler, as it takes into account the context of our job runs and avoids unnecessary scaling. Review collected by and hosted on G2.com.

I believe there is potential to further enhance efficiency by incorporating code optimization, rather than focusing solely on job optimization. Review collected by and hosted on G2.com.

Zipher has consistently delivered measurable Databricks cost efficiencies while improving workload stability. It optimized cluster autoscaling by reducing unnecessary node churn and maintaining a steady balance between on-demand and spot capacity. We saw improved DBU utilization across interactive and job clusters, along with more predictable spend and faster job execution times which leads to reduce our oveall compute cost. Review collected by and hosted on G2.com.

Zipher’s automated optimization works intelligently behind the scenes, analyzing historical Databricks workloads and adjusting resources based on learned patterns. Since much of this logic is abstracted, we don’t always see how decisions are made, but the hands-off experience has been smooth and effective. More visibility into the autoscaling logic would make it even easier to understand and trust the optimizations as they happen. Review collected by and hosted on G2.com.

Zipher performs exactly as promised, providing notable cost savings and enhancing job performance. It helped stabilize our autoscaler by minimizing fluctuations in the cluster node configuration. After a straightforward setup, the platform needs very little ongoing attention, and the support team has been consistently responsive and pleasant to work with. Review collected by and hosted on G2.com.

The primary downside is the absence of built-in dashboards and metrics, which I suspect is due to us being early adopters who have not yet upgraded. However, since we already had adequate monitoring systems to verify the tool's effectiveness, we did not see a reason to proceed with an upgrade. Review collected by and hosted on G2.com.

Extremely easy to use - it’s really a hands-free solution and setup, as Zipher claims. We also were able to reduce our spend on cloud infra and databricks rapidly. Review collected by and hosted on G2.com.

Better dashboards/reporting. At the end of the day, we see the savings on our bills though Review collected by and hosted on G2.com.